大型语言模型 (LLM) 代理越来越多地实现多步骤工程和信息学工作流程的自动化,，但他们很少询问任务实际需要多少工作量。他们通常遵循最大上下文优先策略 - 重新读取他们已经看到的文件和依赖项 - 将一行编辑变成小型代码库审核。我们认为缺失的能力是任务感知执行范围估计:判断任务'的难度,它真正需要的信息,以及提交预算之前的最短可靠路径。我们将最小充足执行和代理认知冗余率 (ACRR), 形式化，并提出 E3 (Estimate, 执行, 扩展): 代理估计初始操作点, 执行最小可行路径, 并仅在验证失败时扩展范围。在 MSE-Bench（能力控制模拟器中进行 121 次编辑的确定性基准）上，E3 匹配最强的基线的 100% 成功，同时将成本削减 85%,、令牌削减 91%,、检查文件 92%,，并进一步击败强大的自适应检索基线 16%;，这些收益在保留的指令措辞和基本上每项成本中都存在。加权。配套的真实模型工具 (LLM-Case) 证实了对实时 gpt-4o 代理编辑真实开源库, 的影响，每个候选补丁都是通过针对测量的 oracle: 实际运行项目的 真实 pytest 套件进行评分的，过度阅读较温和，但真实, 和 E3 是在可比任务成功方面最精简、最快的策略 - 它的一个不足是提供者速率限制, 不是错误编辑。我们将其框架为执行冗余,的受控探测，而不是对任何部署的代理,的测量，并将任务感知执行定位为迈向基于工程的人工智能(EGAI)的一步——代理的努力植根于任务的工程现实。我们发布了框架和基准。
Large language model (LLM) agents increasingly automate multi-step engineering and informatics workflows, yet they rarely ask how much effort a task actually requires. They often follow a maximum-context-first strategy--re-reading files and dependencies they have already seen--turning a one-line edit into a small code-base audit. We argue the missing capability is task-aware execution-scope estimation: judging a task的 difficulty, the information it truly needs, and the shortest reliable path before committing budget. We formalize minimum-sufficient execution and the Agent Cognitive Redundancy Ratio (ACRR), and propose E3 (Estimate, Execute, Expand): the agent estimates an initial operating point, executes a minimum viable path, and expands scope only when verification fails. On MSE-Bench--a deterministic benchmark of 121 edits in a capability-controlled simulator--E3 matches the strongest baseline的 100% success while cutting cost by 85%, tokens by 91%, and inspected files by 92%, and further beats a strong adaptive retrieval baseline by 16%; the gains survive held-out instruction wording and essentially every cost weighting. A companion real-model harness (LLM-Case) corroborates the effect on a live gpt-4o agent editing a real open-source library, with every candidate patch graded by actually running the project的 real pytest suite against a measured oracle: the over-reading is milder but real, and E3 is the leanest and fastest policy at comparable task success--its one shortfall a provider rate-limit, not a wrong edit. We frame this as a controlled probe of execution redundancy, not a measurement of any deployed agent, and position task-aware execution as a step toward engineering-grounded AI (EGAI)--agents whose effort is anchored in the engineering reality of the task. We release the framework and benchmark.
科目: 人工智能 (cs.AI); 计算与语言 (cs.CL); 软件工程 (cs.SE); 系统与控制 (eess.SY)
Subjects: Artificial Intelligence (cs.AI); Computation and Language (cs.CL); Software Engineering (cs.SE); Systems and Control (eess.SY)